Institutional members access full text with Ovid®

Share this article on:

Performance Measures in Neurosurgical Patient Care: Differing Applications of Patient Safety Indicators

Moghavem, Nuriel BA; McDonald, Kathryn MM; Ratliff, John K. MD; Hernandez-Boussard, Tina PhD, MPH, MS

doi: 10.1097/MLR.0000000000000490
Original Articles

Background: Patient Safety Indicators (PSIs) are administratively coded identifiers of potentially preventable adverse events. These indicators are used for multiple purposes, including benchmarking and quality improvement efforts. Baseline PSI evaluation in high-risk surgeries is fundamental to both purposes.

Objective: Determine PSI rates and their impact on other outcomes in patients undergoing cranial neurosurgery compared with other surgeries.

Research Design: The Agency for Healthcare Research and Quality (AHRQ) PSI software was used to flag adverse events and determine risk-adjusted rates (RAR). Regression models were built to assess the association between PSIs and important patient outcomes.

Subjects: We identified cranial neurosurgeries based on International Classification of Diseases, Ninth Revision, Clinical Modification codes in California, Florida, New York, Arkansas, and Mississippi State Inpatient Databases, AHRQ, 2010–2011.

Measures: PSI development, 30-day all-cause readmission, length of stay, hospital costs, and inpatient mortality.

Results: A total of 48,424 neurosurgical patients were identified. Procedure indication was strongly associated with PSI development. The neurosurgical population had significantly higher RAR of most PSIs evaluated compared with other surgical patients. Development of a PSI was strongly associated with increased length of stay and hospital cost and, in certain PSIs, increased inpatient mortality and 30-day readmission.

Conclusions: In this population-based study, certain accountability measures proposed for use as value-based payment modifiers show higher RAR in neurosurgery patients compared with other surgical patients and were subsequently associated with poor outcomes. Our results indicate that for quality improvement efforts, the current AHRQ risk-adjustment models should be viewed in clinically meaningful stratified subgroups: for profiling and pay-for-performance applications, additional factors should be included in the risk-adjustment models. Further evaluation of PSIs in additional high-risk surgeries is needed to better inform the use of these metrics.

Supplemental Digital Content is available in the text.

*Medical School, Stanford School of Medicine

Stanford Center for Health Policy/Center for Primary Care and Outcomes Research

Department of Neurosurgery, Stanford School of Medicine, Stanford University

§Department of Surgery, Stanford School of Medicine, Stanford, CA

Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website,

N.M. was supported by the Stanford School of Medicine Medical Scholars Fund. T.H.-B. was supported by Grant Number K01 HS018558 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

The authors declare no conflict of interest.

Reprints: Tina Hernandez-Boussard, PhD, MPH, MS, Department of Surgery, Stanford School of Medicine, 1070 Arastradero #307, Palo Alto, CA 94305. E-mail:

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.